Jauhar and colleagues’ ^{Reference Jauhar, McKenna, Radua, Fung, Salvador and Laws1} review and meta-analysis of cognitive-behavioural therapy (CBT) for the reduction of particular symptoms associated with schizophrenia is interesting but incomplete. For example, the review does not examine the clinical significance of dose or duration of CBT treatment. This limitation is considerable, as an analysis of effective elements of CBT for psychosis found that ‘consistent delivery of full therapy, including specific cognitive and behavioural techniques, was associated with clinically and statistically significant increases in months in remission, and decreases in psychotic and affective symptoms’, while ‘delivery of partial therapy involving engagement and assessment was not effective’. ^{Reference Dunn, Fowler, Rollinson, Freeman, Kuipers and Smith2}

Jauhar et al have also excluded measurement of long-term outcomes from their analysis, measuring only end-of-study data. This is another considerable limitation, as symptom reductions maintained at 9- or 18-month follow-up represent a substantial benefit of effective CBT. Further, although reduction of psychotic symptoms is an important treatment outcome to measure, CBT is particularly focused on reducing distress associated with such symptoms and improving an individual’s ability to cope with them. As psychotic symptoms can continue even with administration of powerful antipsychotic medication, improvements in these areas may be clinically significant for many CBT recipients. Indeed, a comprehensive synthesis of qualitative research into patients’ experiences of CBT for psychosis ^{Reference Berry and Hayward3} found that the most commonly identified ‘key ingredients’ of CBT included increased understanding of psychosis and of coping strategies, reappraisal of distressing beliefs, and normalisation: ‘Participants did not necessarily experience an actual reduction in the frequency or distressing content of psychotic experiences, but instead gained an increased ability to cope and an increased perception of personal power’. It is also important to consider that not all individuals want their ‘symptoms’ eradicated, and such appraisals are common in the wider literature on recovery from psychosis or schizophrenia: ‘Learning to cope to accept that you hear voices or whatever your symptoms are. Recovery is… to be able to live with it’. ^{Reference Pitt, Kilbride, Nothard, Welford and Morrison4} So, although analyses of CBT that focus only on psychotic symptom reduction are important, they are also incomplete; ‘secondary’ outcomes such as reduced distress or self-defined recovery may be valued more highly than symptom reduction alone by many patients, and such outcomes are increasingly well measured in CBT trials. ^{Reference Greenwood, Sweeney, Williams, Garety, Kuipers and Scott5} Future meta-analyses of CBT will contribute more meaningfully to our understanding of its effectiveness by examining these wider outcome domains and acknowledging their value as long-term benefits.